[MITgcm-support] silica fields in DIC package

[Jonathan Lauderdale]

Dear Hannah,

Thanks for your email last month about silicate in the MITgcm DIC package - we discussed the issue at a dev-team meeting, and I wanted to summarize a few of the issues mentioned:

First, you are absolutely right that: 
(a) the current "sillev1.bin” file has incorrect dimensions, 128x64x15, instead of the expected 128x64x12. As far as I know, MITgcm doesn’t actually ever read the 13th, 14th, and 15th slices but it’s definitely confusing.
(b) the more dynamic carbonate S/R “calculate_saturation” keeps the silicate concentration fixed, although it’s good to hear that you found that the depth of calcite saturation horizon didn’t change much.

I think this corner of the code is not currently part of the nightly testing regime (as you found out previously with the bugs in the Munhoven ph solver) but the team are interested in adding a test case. There are a few things to consider first, and we would appreciate your input:

(1) Temporal variability - there is a slight difficulty with supplying a monthly full-depth Si climatology in that the World Ocean Atlas has full depth (5500m) values for the annual climatology, but the monthly climatology only goes down to 800m depth. Since the lysocline/carbonate compensation depths are going to be sensitive largely below 800m, we could consider supplying the annual field instead. Is this what you found in your testing?

(2) Z-level orientation - the “sillev1.bin” file is for the actual sea surface (0m) I think, because we are calculating air-sea fluxes across the interface, whereas the 3d file should probably be located at cell centers in the vertical (ie first level at 25m), so we’d likely need to supply two files, one for air-sea fluxes, and one for carbonate lysocline/compensation. 

I think the next step would be to raise an issue on GitHub, and we can continue the discussion there. It sounds like you have already made a lot of progress with testing and code, so we would be interested to incorporate your changes via a follow-up pull request.  Here’s a link <https://www.dropbox.com/s/63va1hfap9o0744/silica3d.bin?dl=1> to a silicate file for the tutorial_global_oce_biogeo experiment that you asked about (128x64x15) that should work, if are still willing to run your new code in a few verification simulations and share with us your results about changes in CCD, etc?

Many thanks,

Jonathan
___________________________________________________________________________
Dr. Jonathan M. Lauderdale
Research Scientist
Department of Earth, Atmosphere and Planetary Sciences 
Massachusetts Institute of Technology 
77 Massachusetts Avenue (54-1518)
Cambridge, MA 02139, USA 

Email: mailto:jml1 at mit.edu
Web: http://paocweb.mit.edu/people/jml1
Twitter: https://twitter.com/jon_lauderdale
Blog: https://seamanticscience.wordpress.com/
Git: https://github.com/seamanticscience
___________________________________________________________________________

> On Dec 2, 2021, at 06:07, Hannah Kleppin <hkleppin at marum.de> wrote:
> 
> Dear MITgcm users and developers,
> 
> since some time I have a slightly uneasy feeling every time I stumble across the variable silicaTEST in calcite_saturation.F (line 110) and the comment below "Cmick - TEMPORARY!!!!!".
> I finally decided to give it a closer look and I think the model only reads in surface silica fields (CALL READ_REC_XY_RS <https://lxr.mitgcm.org/lxr2/ident/MITgcm?_i=READ_REC_XY_RS>( DIC_silicaFile <https://lxr.mitgcm.org/lxr2/ident/MITgcm?_i=DIC_silicaFile>,silica0 <https://lxr.mitgcm.org/lxr2/ident/MITgcm?_i=silica0>,intime0 <https://lxr.mitgcm.org/lxr2/ident/MITgcm?_i=intime0>,myIter <https://lxr.mitgcm.org/lxr2/ident/MITgcm?_i=myIter>,myThid <https://lxr.mitgcm.org/lxr2/ident/MITgcm?_i=myThid>), Line 101 in dic_fields_load.F), the silica values used for pH calculation at depth are in anyway fixed to 0.03 mol m^-3. 
> The dimensions of the silica input file in the verification example (https://github.com/MITgcm/MITgcm/tree/master/verification/tutorial_global_oce_biogeo <https://github.com/MITgcm/MITgcm/tree/master/verification/tutorial_global_oce_biogeo>) give the impression that a 3D silica field is used (it has dimensions Nx*Ny*Nr), eventhough if you look at the data, you can see that the lowest 3 levels are empty and the 12 levels above are always surface levels, thus its a silica surface monthly climatology.
> For my own set-up I modified slightly some routines to read in 3d-fields of silica and use them in the calculation of the pH value. The resulting changes in, for example, the depth of the calcite saturation horizon are indeed small. For the tutorial example I cannot do this test since there are only the surface silica fields provided. 
> If you think its worth it I can upload the changes I did in github. But I'm not so familiar with the IO-routines and the exchanges for the overlap regions, so maybe someone would need to have a look on the changes I did. If someone has the deeper silica data of the tutorial example, then I could also run a test with them.
> 
> Cheers,
> Hannah 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.mitgcm.org/pipermail/mitgcm-support/attachments/20220111/fe94a386/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 1864 bytes
Desc: not available
URL: <http://mailman.mitgcm.org/pipermail/mitgcm-support/attachments/20220111/fe94a386/attachment.p7s>